Bibliographic record
Abstract
Information and Communications Technology (ICT) is a large and growing contributor to the world’s GHG emissions. Left unchecked, annual growth in this sector is expected to continue at 12%, as more and more computing resources and storage are required to support the move to a digital economy and the dematerialization of goods and services. Implicit in this growth is the additional demands for energy to run the storage and computing power systems (data centres). The challenge becomes one of meeting the increasing demand for technology, information storage, and computing power, while reducing the overall impact on the environment. CANARIE has recognized this challenge and is sponsoring research into the opportunities for green or carbon-neutral computing. This report seeks to address two questions: • How can universities and other organizations create and maintain carbon neutral computing facilities that are cost-competitive and scalable? • Should an organization or user move its servers (co-located or rented) to a low-carbon region, and can that move be financed through carbon offsets? Six alternative scenarios for creating a carbon-neutral data centre were developed and evaluated using financial analysis and Net Present Value (NPV), as well as a life cycle assessment and GHG audit to factor in materials, construction, and use-phase emissions. Two key drivers emerged as critical factors for creating low-carbon computing; location and cost. Location - The electricity generation mix in Canada varies considerably from one province to the next, and the dirtiest region (Alberta) is over 70 times as GHG intense as the cleanest (Quebec). Location is therefore a key determinant of how clean or dirty a data centre’s GHG emissions profile is. Data centres located in regions with access to renewable energy provide opportunities for the data centre to tap into the clean energy source, either by connecting to the regional electricity grid or through direct investment in renewable energy projects. Cost - Data centres and the required electrical supply and infrastructure constitute a significant capital investment. Large data centres require a significant amount of electricity, and can therefore justify investments in dedicated renewable energy projects and other forms of direct investment to secure clean electricity. Small data centres, including those evaluated in this report, are less able to justify and support large investments in renewable energy. These owners should therefore leverage indirect investments in clean electricity to achieve carbon-neutrality, be it through Renewable Energy Credits (RECs) or premiums, or carbon offsets. The recommendation, aimed at small and medium sized data centres, is to build in low-carbon regions where opportunities exist to tap into renewable energy sources. Data centres and any renewable energy sources should be connected to the region’s electricity grid in order to manage variability and maximize investments. In this way, investments in renewable energy are spurred on by increasing regional demand while any new sources are connected to the regional grid. The second question deals with existing data centre users (co-located) looking to achieve carbon-neutrality for their operations. Carbon offsets and credits offer opportunities for projects to receive financing to support projects that achieve carbon reductions, however the opportunity to use carbon financing and the decision of whether to move locations must be evaluated on a case-by-case basis. Overall it is recommended that data centres located in low to medium carbon-intensity regions should simply invest in RECs or carbon offsets rather than financing a relocation through carbon credits. Only those data centres that are currently located in dirty or high-carbon intensity areas have a significant opportunity to capitalize on carbon credit financing.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.011 | 0.035 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.004 | 0.004 |
| Scholarly communication | 0.015 | 0.008 |
| Open science | 0.003 | 0.004 |
| Research integrity | 0.005 | 0.004 |
| Insufficient payload (model declined to judge) | 0.012 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".